Nodes/Nougan Nodes/Nougan MiniMax H3 KSampler 🎬
ComfyUI Node

Nougan MiniMax H3 KSampler 🎬

MiniMax H3 sampler: SamplerCustomAdvanced-shaped (noise / guider / sigmas / latent in), CFG-free like H3 itself, with live preview and VAE + audio-VAE decode built in. model/positive are only needed to auto-build a BasicGuider when no guider is wired.

By WinnouganΒ·Created 4 months agoΒ·Updated 3 days agoΒ· 31
Nougan MiniMax H3 KSampler 🎬
  • latent_image
  • noise
  • guider
  • sampler
  • sigmas
  • model
  • positive
  • vae
  • audio_vae
  • latent
  • denoised
  • image
  • audio
β—„sampler_nameeulerβ–Ί
β—„schedulerbetaβ–Ί
β—„steps8β–Ί
β—„steps_to_run-1β–Ί
β—„denoise1.00β–Ί
β—„seed0β–Ί
β—„decodetrueβ–Ί
β—„tiled_decodefalseβ–Ί
β—„preview_fps12β–Ί
β—„previewtrueβ–Ί
β—„preview_max_res1024β–Ί
β—„preview_frames1β–Ί
β—„preview_jpeg_quality80β–Ί
β—„preview_tiny_vaenoneβ–Ί
CategoryNougan/sampling

Inputs (23)

NameTypeDefaultDescription
latent_imageLATENTβ€”
sampler_nameCOMBOeulerBuilt-in ComfyUI samplers (same list as KSamplerSelect).
schedulerCOMBObeta'none' = the schedule comes from the wired sigmas input.
stepsINT81–10000H3 is distilled: the reference workflow uses 8 steps.
steps_to_runINT-1-1–10000Chainsampling: run only the FIRST N steps of the schedule, then stop partially denoised (a second sampler continues from there). -1 = all steps. Also slices a wired sigmas schedule.
denoiseFLOAT1.000–1β€”
seedINT00–18446744073709550000Ignored when a NOISE input is connected.
decodeBOOLEANtrueDecode the result with the connected VAE(s). Off = latent only.
tiled_decodeBOOLEANfalseβ€”
preview_fpsINT121–60Playback FPS of the live preview (ignored for clips with audio).
noiseoptNOISECustom noise source (e.g. RandomNoise). Replaces the seed widget when connected (the widget hides itself).
guideroptGUIDERCustom guider (e.g. BasicGuider). Carries the model and conditioning, so model/positive are optional with one wired.
sampleroptSAMPLERCustom sampler (e.g. KSamplerSelect). Replaces the sampler_name widget when connected (the widget hides itself).
sigmasoptSIGMASCustom sigma schedule (e.g. BasicScheduler). Replaces the scheduler / steps / denoise widgets when connected (they hide themselves).
modeloptMODELUsed to build a positive-only BasicGuider when no guider is wired (and for sigmas + the live preview). Optional when a guider is wired.
positiveoptCONDITIONINGUsed with model to build the BasicGuider. Optional when a guider is wired.
vaeoptVAEVideo VAE: decodes the result into the IMAGE output and can drive the preview.
audio_vaeoptVAEAudio VAE: decodes the audio half of H3's AV latent into the AUDIO output and drives the spectrogram preview.
previewoptBOOLEANtrueβ€”
preview_max_resoptINT10240–8192β€”
preview_framesoptINT11–1024Frames per step for the animated video preview (1 = single frame).
preview_jpeg_qualityoptINT8030–100β€”
preview_tiny_vaeoptCOMBOnoneTiny VAE decoder (models/vae_approx) for true-RGB previews, like KJ's Model Preview Override. Needs the decoder code: vendored tiny_vae.py or a KJNodes install.

Outputs (4)

NameTypeDescription
latentLATENTβ€”
denoisedLATENTβ€”
imageIMAGEβ€”
audioAUDIOβ€”